1. Field of the Invention
The present invention relates to photodetectors. More particularly, the present invention relates to photodetection systems employing photomultiplier tubes.
2. The Prior Art
Optical sensing is the process of converting optical signals (photons) into electrical signals (electrons). In most applications, where the optical signals are large, or the temporal frequencies are low, this is done using solid state devices known as Photodiodes. Photodiodes are inexpensive (typically $1 to $30) and simple to use. They have a high dynamic range, and can be very fast when the light levels are large.
For signals which are too dim to see, photodiodes become much slower, due to the relatively high capacitance of the diode and the small currents generated by the dim signals. Even though photodiodes have excellent dynamic range (they are linear over up to 14 orders of magnitude) their output is proportional to the optical signal, so in practice their useful dynamic range is quickly limited by subsequent electronics. (high speed analog to digital converters typically have a dynamic range of only 3.6 orders of magnitude).
When the optical signals are dim, and frequencies are high, photomultiplier tubes are typically used. Photomultiplier tubes are expensive (typically $100 to over $1000), are subject to damage from overloading input signals, such as ordinary room light, and require a high voltage (typically 1000 Volts) to operate. A photomultiplier tube consists of a photocathode, one or more dynodes (typically 10) and an anode. Individual photons striking the cathode have up to a 25% chance of dislodging an electron. These photoelectrons are then accelerated towards the first dynode by an electric field. When they strike the dynode they dislodge additional electrons, thus amplifying the signal. These secondary electrons then cascade towards the next dynode where they are again amplified. At the end of the dynode chain, the electrons are collected by the anode which carries them outside of the photomultiplier tube. At this point, the signal is large enough to be easily measured using conventional electronics, such as a transimpedence amplifier, followed by an analog-to-digital converter.
The gain at each dynode is a function of the energy of the incoming electron, which is proportional to the electric potential between that dynode and the previous stage. This relation is of the form: EQU G(v)=k.times.v.sup.a
where a is typically in the range of 0.6 to 0.8. The total gain of the tube is the product of the gains from all the dynodes. Typically, and as shown in FIG. 1, the bias voltages for the dynodes are generated by connecting a string of voltage-divider resistors between the cathode, all the dynodes, and ground. Typically the resistance, and therefore the voltage, between all of the dynodes and between the last dynode and anode are the same. This resistance, R, is used as a scaling constant. Typically, the resistance between the anode and the first dynode is 1.5R to 3.5R where R is usually between 10K and 100K ohms. A large negative voltage is then applied to the cathode, and the potential is divided up evenly across the dynodes by the voltage-divider resistor chain.
This conventional biasing scheme is useful for operating the photomultiplier tube at a single programmable gain. Altering the applied cathode voltage changes the gain according to the relation: EQU G(v)=k.times.v.sup.a'
where a' is typically in the range of 7.0 to 8.0. However, the large voltages involved make it difficult to change the gain quickly, due to parasitic capacitances and the large resistor values needed to limit power dissipation in the bias string. The conventional usage is to decide on a tube gain in advance, set the appropriate cathode voltage and then operate the tube at that voltage throughout the measurement operation.
In this configuration, the dynamic range of the photomultiplier tube is limited on the low end by the noise and gain characteristics of the transimpedence amplifier and, on the high end, by the ability of the photomultiplier tube to deliver anode current. The anode current is limited by space charge effects within the tube, by bias string power consumption, and by the consumable nature of the material coating the dynodes. If the optical signals to be measured are short pulses with low duty cycles, then capacitors can be placed across the last few bias resistors to improve pulse linearity. However, this trick does not help for bright signals which have a high duty-cycle, or which last more than a few tens of microseconds.
Logarithmic compression has been employed in the prior art in applications such as optical densitometers. These are devices used by photographic labs to check the darkness of an exposed and processed piece of film. However, they only operate over time scales of milliseconds to seconds. Examples of the use of this technique are found in U.S. Pat. Nos. 3,586,443, 3,653,763, and 3,733,491.
Additional examples of photomultiplier tube gain modulation appear in a number of forms, including methods for pre-setting the gains of various tubes which share a common HV supply, AGC circuits for making qualitative measurements, photomultiplier tube overload protection circuits, and fast gain switching circuits.